# facebookresearch/dlrm

An implementation of a deep learning recommendation model (DLRM)

Repository: https://github.com/facebookresearch/dlrm
Canonical: https://ross.abutalabs.com/products/dlrm
Language: Python
License: MIT
License Family: permissive
Last push: 2026-01-12T11:07:16+00:00

## Health v2 (maintenance only)
Score: 60/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 62, release rhythm 35, longevity 100
- inputs: {"age_days": 2669, "days_push": 233, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4065, forks 860 (observed 2026-08-28T04:08:34.100057+00:00)

## What it is
A PyTorch-based implementation of the Deep Learning Recommendation Model (DLRM) from Facebook Research, which processes dense and sparse features through MLPs, embedding tables, and feature interaction operators to predict click-through probability. It serves as both a reference implementation of the DLRM architecture and a benchmark for recommendation system hardware and software.

## Use cases
- implement a deep learning recommendation model
- train a click-through rate prediction model
- benchmark recommendation model performance on GPUs
- learn how sparse features and embedding tables work in recommenders
- reproduce the DLRM paper results
- prototype personalized recommendation systems

## When to choose
- you need a reference implementation of the DLRM architecture
- you're benchmarking recommendation workloads on GPUs or CPUs
- you want to experiment with dense/sparse feature interaction models
- you're doing research on recommendation systems

## When to avoid
- you need a production-ready, full-featured recommendation platform
- you want a non-PyTorch framework like TensorFlow
- you need collaborative filtering out of the box rather than a neural model
- you're looking for a hosted recommendation service

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, benchmarking
- domain: machine-learning, deep-learning
- platform: python, cross-platform
- tags: recommendation-model, dlrm, pytorch, ctr-prediction, embedding-tables, facebook-research, recommendation-systems, gpu, linux

## Member repositories
- facebookresearch/dlrm (main) score 60

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:34.100057+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:23:33.440787+00:00, confidence not recorded.
  - readme: https://github.com/facebookresearch/dlrm (fetched 2026-08-28T04:08:34.100057+00:00, sha 83fd1cb41f52)
- Data as of 2026-08-30T08:39:29.467469+00:00.